Authors:

Karla Dominguez

Date Created:

Not specified

Course Title:

Professor:

Not specified

About Paper:

This horticultural research project seeks to establish a dedicated testbed for image data collection and analysis in order to improve crop performance comprehension and forecasting. The focus of the project is the management of a small research plot devoted to the cultivation of tomato and watermelon varieties, which provides an environment for imaging experiments. The primary objective is to create a pipeline for image data collection that captures LIDAR and RGB images of each crop throughout the growing season. This pipeline will facilitate the creation of a comprehensive image database, which will be used for the development of sophisticated models. An algorithm for image data processing will be devised to extract necessary variables, such as plant volume, for calibrating DSSAT crop models and machine learning models. These models will facilitate variety-specific yield prediction and corresponding analytic capabilities. The project will expand the team's knowledge of agricultural research plot management, including crop cultivation and maintenance. Using computer vision techniques, the team will acquire proficiency in developing Python scripts for image data processing. Extraction and transformation of data from LIDAR and RGB images will facilitate subsequent analysis. Additionally, the initiative will provide a unique opportunity to investigate the application of computer vision in the agricultural sector. The team seeks to improve crop monitoring, disease detection, and yield forecasting by utilizing advanced image analysis techniques and the incorporation of crop models. Aside from contributing to the field of agricultural informatics, this initiative will provide valuable insights for sustainable and effective crop management practices.

Source:

Purdue University / 2023

Topics:

No topics listed

Co-authors:

Karla Dominguez